Also modify test code to hopefully avoid deadlock on failure.

A pthread_atfork() child handler is used to seeded the PRNG with pid, time
and some stack data.

A pthread_atfork() child handler is used to seeded the PRNG with pid, time
and some stack data.

Issue 18772:  Restore set dummy object back to unicode and restore the identity checks in lookkey().

The Gdb prettyprint plugin depended on the dummy object being displayable.
Other solutions besides a unicode object are possible.  For now, get it
back up and running.

The identity checks in lookkey() need to be there to prevent the dummy
object from leaking through Py_RichCompareBool() into user code in the
rare circumstance where the dummy's hash value exactly matches the hash
value of the actual key being looked up.

RuntimeError exception.  Patch by Christian Schubert.

RuntimeError exception.  Patch by Christian Schubert.

Patch by Yogesh Chaudhari.

Patch by Yogesh Chaudhari.

Issue #7732: try to fix test_bug7732's flakiness on Windows by executing it in a fresh temporary directory.

Patch by Meador Inge.

See also issue #18647.

Patch by Meador Inge.

See also issue #18647.

determined minimal width of repeated subexpression is >0 (an empty line was
not matched in any case).

determined minimal width of repeated subexpression is >0 (an empty line was
not matched in any case).

Now max width is MAXREPEAT on 32- and 64-bit platforms when one of
subexpressions is unbounded repetition.

Now max width is MAXREPEAT on 32- and 64-bit platforms when one of
subexpressions is unbounded repetition.

OpenSSL 1.0.0+ instead of the deprecated CRYPTO id callback function.

OpenSSL 1.0.0+ instead of the deprecated CRYPTO id callback function.

(original patch by Phil Webster).

- document abitype.py, analyze_dxp.py, get-remote-certificate.py,
  import_diagnostics.py, parse_html5_entities.py.
- remove redemo.py which was moved to Tools/Demo.
- sort list into alphabetical order

Patch by Seydou Dia.

* describe how \w is different when used in bytes and Unicode patterns.

* describe re.ASCII flag to change that behaviour.

* remove personal references ('I generally prefer...')

* add some more links to the re module in the library reference

* various small edits and re-wording.

(trying to fix a failure on the FreeBSD 9.0 buildbot)